anthropic/claude-fable-5
The June 2026 Fable model, succeeded by Fable 5.1 — Mythos-class, above Opus.
Tune the request, see the current price ceiling, and inspect the real response without leaving this model page.
Sign in first to upload images to this vision model.
Total: 2.2 billion tokens per day
Here's the calculation:
Calls per day:
Tokens per call:
Total tokens per day:
Breaking it down by type:
Claude Fable 5 was released by Anthropic on June 9, 2026, alongside Claude Mythos 5, which is the same underlying model with safeguards lifted in some areas and is open only to verified organizations. Anthropic calls Fable 5 a Mythos-class model made safe for general use, and defines Mythos-class as a tier above Opus in capability; at launch it said Fable 5's lead over its other models grows with task length and complexity. It became legacy when Fable 5.1 arrived on September 1, 2026, with retirement not sooner than June 9, 2027.
It has a 1M-token context window, up to 128K output tokens per request, image input and tool calling, a reliable knowledge cutoff of January 2026, and the tokenizer introduced with Opus 4.7. Adaptive thinking is always on: thinking "disabled", manual budgets and assistant prefill all return a 400. Anthropic also rejects non-default temperature, top_p and top_k; RouteMux strips them before forwarding. Effort defaults to high, and Anthropic notes that Fable 5 still performs well at lower levels, often beating the xhigh results of earlier models.
Compared with Fable 5.1, Fable 5 has fewer restrictions. 5.1 rejects forced tool use (tool_choice "any" or a named tool); 5.1 checks whether the system prompt, tools or earlier messages before a thinking block were changed, and can error if they were; and in long agent loops 5.1 may issue one tool call per turn where Fable 5 batches several. Integrations that rely on any of that can stay pinned to Fable 5. Per-message effort, turn-scoped system messages and text watermarking are 5.1 additions, and cache reads are priced differently on the two.
Through RouteMux, Fable 5 costs $3 per 1M input tokens against Anthropic's $10, $15 per 1M output tokens against $50, and $0.3 per 1M cache-read tokens against $1.
Prices verified 2026-10-02
Cache read $0.3/M · Cache write $3.75/M
tool_choice "any" or a named tool is honored. Fable 5.1 returns a 400.budget_tokens is not accepted. Effort defaults to high across five levels, with xhigh for the hardest tasks.Limited time10% extra Credits on your first top-up — up to 200 bonus Credits
Applies to your first top-up within 72 hours of signing up. Minimum $20, once per account.
Use any of these IDs to call this model via the API.
anthropic/claude-fable-5Replace the ROUTEMUX_KEY placeholder with your API key. Create one →
from anthropic import Anthropic
client = Anthropic(
base_url="https://api.routemux.com/anthropic",
api_key="<ROUTEMUX_KEY>",
)
message = client.messages.create(
model="anthropic/claude-fable-5",
max_tokens=1024,
messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(message.content[0].text)The June 2026 Fable model, succeeded by Fable 5.1 — Mythos-class, above Opus.
Claude Fable 5 supports a context window of up to 1M tokens, with up to 128K output tokens per request.
Through RouteMux, Claude Fable 5 costs $3 for input and $15 for output (USD per 1M tokens), below the official list price.
Claude Fable 5 is OpenAI-compatible: point your base URL at RouteMux and set the model field to anthropic/claude-fable-5 — no code changes needed.
Claude Fable 5 accepts text, image as input and produces text as output.
Claude Fable 5 was released on June 9, 2026.
Reviewed by the RouteMux team on 2026-10-24
101 requests · last 30 days